Chapter #

POSITIVISM OR NON-POSITIVISM - TERTIUM NON DATUR A Critique of Ontological Syncretism in IS Research

Bernd Carsten Stahl Centre for Computing and Social Responsibilty, De Montfort University, Leicester LE1 9BH, UK, bstahl@dmu.ac.uk

Abstract: This paper revisits the debate between positivism and its alternatives in the field of information systems from a philosophical point of view. It will argue that the heart of the debate is the ontological difference between the views of reality as observer-independent versus observer-dependent. The logical axiom of the excluded third (tertium non datur) informs us that two contradictory options cannot simultaneously be true. The paper will discuss what the incompatibility of the ontological positions of positivism and its alternatives means for IS research. It will discuss why scholars attempt to mix the two and will spell out the consequences of an acceptance of their incompatibility. The paper will end by arguing that this debate needs to be contextualised with the problem of positivism versus non-positivism in society and it will ask whether a tolerant coexistence of the two approaches is feasible. Without this contextualised understanding of ontology in general, regional ontologies in IS are not likely to be successful as they will be based on unclear bases.

Key words: positivism, non-positivism, interpretivism, ontology, epistemology, methodology

1. INTRODUCTION

"In my view, it is time to assign the rhetoric of positivism versus interpretivism to the scrap heap. It no longer serves a useful purpose. On the contrary, it promotes unhelpful schisms among scholars." (Ron Weber, Editor-in-Chief, MISQ (2004, p. xi))

2 Chapter #

Many information systems (IS) scholars probably sympathise with Weber’s sentiment and agree that they have heard enough of the discussion of positivism versus interpretivism. There seems to be little progress in the debate and it has the potential to hurt the discipline. This paper will nevertheless address the issue of positivism and its alternatives as philosophical foundations of IS research. The reason for this is that there seems to be a growing tendency to mix positivist and non-positivist research. The paper will argue that such philosophical syncretism is suspicious for logical reasons.

The argument presented here is a philosophical one. First, the main argument is based on the ontological assumptions of research. Ontology is one of the classical sub-disciplines of philosophy. Second, it will argue that, in order to understand the current state of philosophical syncretism of IS research, one needs to look at the relationship between ontology, epistemology, and methodology, which opens the doors to other philosophical areas. Third, the central problem discussed here is a conceptual one. "Philosophy has always been concerned with conceptual clarifications" (Tugendhat, 1992 p. 113; cf. Ricoeur, 1995; Wittgenstein, 1963). Finally, the heart of the argument, the tertium non datur, the proposition of the excluded third, is an integral part of Aristotelian propositional logic.

It is generally accepted that research is based on several interconnected philosophical assumptions. This paper will follow Davies (1991) in concentrating on the relationship of ontology (theory of being / reality / essence), epistemology (theory of knowledge), and methodology (theory of method / action). The question of which research philosophy or approach is appropriate in IS has a long history (cf. Pettigrew, 1985) but is not likely to be resolved (Petter & Gallivan, 2004). The ontological divide between different research philosophies has been a continuing topic of debate in the philosophy of science and can be traced back to the great philosophers (cf. Monod, 2002) and even to the earliest writings on western philosophy. The aim of the paper is therefore not to "solve" the problem, which may be impossible (Fitzgerald & Howcroft, 1998), but to clarify its roots and meaning and to spell out the theoretical and practical consequences of the dichotomy between positivism and non-positivism.

The paper thus aims to clarify some questions of general philosophical ontology in the area of information systems. One should note that such a general ontology is of central importance if one wants to discuss regional ontologies, as suggested by Kishore, Sharman & Ramesh (2004). Regional ontologies and the universes of discourse they are based upon rely on general ontology. The very idea that there are regional universes of discourse and that these allow the classification of specific ontological structures is

#. Positivism or Non-Positivism - Tertium Non Datur 3 already a strong ontological claim, which, as the present paper will argue, is likely to be contentious. The dichotomy between positivism and non-positivism is therefore of decisive importance for most questions of ontology in IS.

1.1 Contribution

This paper is written from a non-positivist point of view. The author was motivated by the intrinsic contradictions created by attempts to mix positivism and alternative ontological positions. The paper is positioned outside the traditional ways of doing research in IS. Leading scholars of information systems suggest that researchers following a non-traditional (meaning non-positivist) approach should explain their research to the reader and provide criteria for evaluating it (Klein & Myers, 2001; Lee, 2001). Accordingly, Klein & Myers (1999) suggest a set of criteria for evaluating interpretive research. Similar criteria for philosophical / conceptual research in IS do not exist. Some criteria can nevertheless be distilled from the aims of philosophical research and the overlap it has with other IS research in creating accessible knowledge.

Jenkins (1985, p.112) defines philosophical research as a mental pursuit where the "researcher thinks and logically reasons causal relationships". Logical stringency will thus be one criterion for the evaluation of this paper. The paper should furthermore address a topic that "should be directly related to the future interest that key stakeholders […] are likely to hold in a topic" (Benbasat & Zmud, 1999 p. 8). The argument should be convincing to the reader, it should ring true, be persuasive, and "convince the reader that a believable story is being told" (Trauth & Jessup, 2000 p. 68). Since the topic of the paper concerns the theoretical underpinnings of IS research, one might ask what a good theory is (van de Ven, 1989). Weick (1989 p. 517) suggests that a good theory should be "interesting rather than obvious, irrelevant or absurd, obvious in novel way:, a source of unexpected connections, high in narrative rationality, aesthetically pleasing, or correspondent with presumed realities". Maybe the ultimate quality criterion of the paper will be whether the author can "convince and cajole [his] colleagues about the directions we should now pursue" (Benbasat & Weber, 1996 p. 398).

2. ONTOLOGY

Ontology, which etymologically means "speaking of being", is the philosophical discipline that asks "what is?" and "what does it mean to be" (cf. Heidegger, 1993). It researches the fundamental questions of being, and

4 Chapter # thus, in everyday parlance, one could say that it studies the nature of reality. Ontological assumptions form one of the most important building blocks of our worldview and they are so fundamental that we rarely question them. They are therefore of central importance to any research in any discipline (cf. Klein, Hirschheim & Nissen, 1991). One needs to know what is or what exists in order to research it. There are profoundly different ontological theories, which in this paper will be called ontologies used by different individuals. This paper will argue that the concept of positivism can be defined primarily in terms of its ontological assumptions and that ontological questions are at the basis of many of our epistemological and methodological differences. There are numerous questions of importance to be found in the history of ontology. The paper will concentrate only on the question whether reality is independent of the observer or not.

2.1 Positivism

I will argue in this paper that positivism can be defined as a research approach that is based on the ontological doctrine that reality is independent of the observer. Most scholars interested in the philosophy of IS research agree with this definition. The independent and objective existence of reality can be found as a definition of positivism in a number of texts (cf. Orlikowski & Baroudi 1991; Visala, 1991; Jönsson, 1991; Landry & Banville, 1992; Darke, Shanks & Broadbent, 1998; Iivari, Hirschheim & Klein, 1998; Myers & Avison, 2002; Varey, Wood-Harper & Wood, 2002). Some authors use different terms to denote this ontological position, such as "objectivism" or "realism" (Burrell & Morgan, 1979; Hirschheim, 1985; Chua, 1986; Hirschheim & Klein, 1989; Weber, 2003). These authors typically see positivism as comprising epistemological (Olaison, 1991; Lee, 1991; Walsham, 1995), methodological (Benbasat & Weber, 1996), and sometimes other philosophical aspects, such as ethics (Wynn, 2001). Such a collection of different philosophical aspects under the term "positivism" is understandable for several reasons and some of these aspects will be discussed later.

The division between positivism and non-positivism is so deep that the very term "positivism" has taken on a pejorative meaning for non-positivist (Burrell & Morgan 1979; Lee 2001). It seems to be used predominantly by opponents of the idea, whereas proponents prefer the adjective "positive" instead of "positivist" (cf. Friedman 1994).

The realist ontology on which positivism is based, namely the conviction that reality exists independent of the observer, is a remnant of the successful ontology of the mechanistic natural sciences of the Enlightenment period. It is still wide-spread in parts of natural and technical sciences. Our current use

#. Positivism or Non-Positivism - Tertium Non Datur 5 of the term is shaped by the attempt to import this scientific ontology into humanities and social sciences with the declared aim of copying the success of the natural sciences (cf. Habermas, 1974).

It is not the purpose of this paper to discuss the shortcomings of positivism. Intellectual honesty requires nevertheless that some of the main arguments against positivism are reviewed. This will also help the reader comprehend the thrust of the argument. Positivism, especially the elaborate set of theories developed by the Vienna Circle called "logical positivism", has largely been discredited in the philosophy of sciences. It nevertheless continues to be a strong "logic in use" (Landry & Banville, 1992) or "ontology in use" (Lee, 2004) in the social sciences, and arguably in information systems. The probably most important critique of positivism in information systems is that the idea of an independent reality and the consequent impartial observation of this reality does not lead to an adequate understanding of the phenomena in question (Adam, 2001). This often implies a distinction between natural and social reality. Positivism seems to be a deficient basis for research at least in the latter realm (Nissen, 1985; Orlikowski & Baroudi 1991). Other points of critique address the epistemological problems resulting from a positivist ontology. There are problems of induction and generalisation (Pettigrew, 1985; Lee & Baskerville, 2003). Philosophy has not found a convincing explanation how the mind can adequately represent a mind-independent reality (Khlentzos, 2004). Then there is the charge that positivism is self-contradictory because it is not itself a natural occurrence independent of the observer and related problems of the final foundation of positivism (Quine, 1980). Resulting from this it can be argued that positivism is structurally circular and that despite its alleged objectivity it can only investigate phenomena that are created by the investigator (Stahl 2003).

2.2 Non-Positivism

If positivism is not the only ontological position, then the question arises which alternatives to positivism there are. In the light of our definition of positivism, the alternatives are those ontological positions that do not depand on a reality independent of the observer. The history of philosophy offers a number of different non-positivist ontological viewpoints. Since the observer plays a part in the constitution of reality and is usually believed to do this using his or her mind, some authors hold that the opposite of positivism is rationalism (Hollis, 1994). But the field of rationalism is wide and contains many different theories. One is the doctrine that the observer constructs reality and that, at the extreme, all of reality is just a figment of the individual’s imagination. This solipsist theory is mirrored in the doctrine of

6 Chapter # radical constructivism (Glasersfeld, von, 2000; Watzlawik, 2001). Another non-positivist stream of thought that was hugely influential in the philosophy of the 19th and 20th century is idealism, or more specifically German idealism. It is linked with the names of Hegel, Schelling, Fichte, and others who argued for the supremacy of the mind or spirit over any external reality.

In contemporary IS research these ontological positions do not play a major role. They can be seen, however, as the basis of some of the ontological alternatives to positivism prevalent today. The two most important concepts opposed to positivism are constructionism and interpretivism. Constructionism (or social constructivism) holds that reality is constructed by the observer, but, in opposition to (radical) constructivism, it states that reality is a collective construction. It emphasises the role of interaction and communication in the process of constructing reality (Gergen, 1999). Its intellectual history can be traced back to idealism (Burrell & Morgan 1979). IS researchers who subscribe to the constructionist ontology typically call themselves interpretivists rather than constructionists. The difference between constructionism and interpretivism appears to be that constructionists are more radical and extend their ontological views to all aspects of reality whereas interpretivists limit it to social reality. Since IS researchers are usually interested in aspects of technology having to do with social phenomena, they can mostly refrain from defending the more contentious claims of constructionism and concentrate on those aspects of reality that are easily recognisable as socially constituted.

In current IS research interpretivism is probably the most important alternative to positivism. This paper argues that this is possible because the heart of the interpretivist position is an ontological position which views reality as a social construct. The constructionist ontology of interpretivism can be found in many of the texts on interpretivism (Orlikowski & Baroudi 1991; Walsham, 1995; Darke, Shanks & Broadbent, 1998; Varey, Wood-Harper & Wood, 2002). The concept of interpretivism has to contend with some of the same problems as positivism in that it often refers to epistemological (Myers & Avison 2002; Klein & Myers 2001; Trauth & Jessup, 2000; Walsham, 1995b) and methodological (Lee 1991; Yin, 2003) aspects of research simultaneously. The term "interpretivism" is often not sharply defined. Also, "interpretivism" is a term that is relatively new but at the same time ubiquitous among non-positivist IS researchers. A brief comparison of the uses of "positivism" and "interpretivism" may shed some light on this. If one compares the proceedings of the IFIP WG 8.2 conferences 1984 (Mumford et al. 1985) and 2002 (Wynn et al. 2003) then one finds a marked difference between use and acceptance of these terms. In 1984 the main thrust of the conference was to break the perceived

#. Positivism or Non-Positivism - Tertium Non Datur 7 stranglehold of positivism on IS research. Consequently, terms starting with "positivi…" were used 158 times in 9 of the 18 essays. "Interpretiv" was used only 15 times, mostly with references to Burrell & Morgan (1979). "Interpretivism" or "interpretivist" were not used at all. In 2002, when non-positivist IS research was firmly established, "interpretive" was used 30 times in 10 out of 30 essays whereas "positivi…" was used only twice, one of which was a reference. This indicates that the struggle against positivism seems to have been successful and interpretivism is recognised as a term in IS research.

2.3 Tertium Non Datur

The paper has so far been set up to sharpen the contradiction between positivism and non-positivism. It has defined positivism as the ontological claim that reality is independent of the observer and non-positivism as the logically contradictory view that reality depends on the observer. The advantage of this definition is that it allows the application of a fundamental logical axiom, namely the proposition of the excluded third. This proposition was developed by Aristotle in his Metaphysics. It means that a sentence must be true or false. In the notation of propositional logic it can be represented as follows:

¬(p � ¬p) The history of logic has given rise to several attempts to show that this

axiom is not sufficient and that logical states do not have to be bivalent. Examples of such non-bivalent logics are modal logic, deontic logic (Garson, 2003) or fuzzy logic (Hajek, 2002). The proposition is nevertheless widely accepted and forms one of the basic tenets of our scientific system. An example can easily show the strength of the axiom. If A is the proposition "X is a dog", then ¬A is the proposition that "X is not a dog". Tertium non datur informs us that it is impossible that A and ¬A are true, thus that X can not be a dog and not a dog (or a non-dog: ¬dog) at the same time.

If the proposition A means: "reality is independent of the observer" then ¬A can be translated as "reality is not independent of (thus dependent on) the observer". According to tertium non datur, both cannot be true simultaneously. This statement is the heart of this paper. The irreconcilable opposition between positivism and non-positivism is simply based on a logical axiom and the ontological root of the terms.

This does not solve all ontological problems in IS research but it allows for a much more concise discussion of several issues. We can now say, for example that, if interpretivism is a form of non-positivism (as was argued earlier), then a researcher cannot follow a positivist and an interpretivist

8 Chapter # research approach at the same time. That does not mean that all researchers must be either positivists or interpretivists. Similarly, if a cat is a non-dog then X cannot be a dog and a cat. X does not have to be a cat or a dog, however, since X might be, say, a fish. An analogue conclusion is that an IS researcher can choose a non-positivist ontology that is not interpretivist.

This position should not be equated with some of the theories concerning the relationship between positivism and its alternatives viewed as paradigms, such as purism (Petter & Gallivan, 2004), supremacism (Klein, Hirschheim & Nissen, 1991), or paradigm incommensurability (Brooke, 2002; Mingers, 2001). It only states that the ontological assumptions of positivism and non-positivism are not commensurable. What this means for research epistemology and methodology will be explored in the following sections.

3. EPISTEMOLOGY

Epistemology is the "theory or science of the method or grounds of knowledge" (OED 2004). If research wants to produce knowledge then it has to rely on an implicit or explicit epistemology. This paper will follow Chua (1986) in distinguishing between epistemology and methodology where the former refers to the principles of knowledge, the latter to ways of acquiring it. Epistemology is closely linked to ontology. One can only gain knowledge about entities that exist. In return, one needs to have a way of gaining knowledge in order to make statements whether something exists. While the exact relationship between given ontologies and epistemologies is not always clear-cut, it is important to note that every epistemology requires a corresponding ontology (Iivari, Hirschheim & Klein, 1998). There is no ontology-free epistemology (cf. Feyerabend, 1980).

If epistemology is the philosophical discipline concerned with knowledge then it needs to define what knowledge is. One definition sometimes used by philosophers is that knowledge is "true, justified beliefs" (Steup, 2001). This is helpful because it allows us to distinguish the question when a statement is true from the question when we are justified in believing a statement to be true. While the latter question is much-discussed in IS research, the former, arguably more important one, is usually ignored.

3.1 Truth Theories

The question when a statement is true is addressed by theories of truth. There are at least four major types of such theories using different criteria for determining the truth of a proposition: correspondence, consensus, coherence, and pragmatist. The correspondence theory of truth holds that a

#. Positivism or Non-Positivism - Tertium Non Datur 9 statement is true if it corresponds with the reality it describes. This is probably closest to our everyday understanding of truth (McCarthy, 1992) and assumes an objectively given describable reality. Proponents of onsensus theories believe that truth is defined by the agreement of those who are knowledgeable in the area in question (Wellmer, 1986; Habermas 1991). Coherence theories hold that statements are true if they are can be supported in a formal system of statements. Typical examples of this are mathematical theorems (Weizenbaum, 1976; Wittgenstein, 2001). Pragmatist theories hold that those statements are true that fulfil their purpose, that lead to desirable outcomes (Rorty, 1991).

These different theories of truth are of highest importance for any research, including IS research, because they determine what can count as a successful attempt to produce truth and thus knowledge. They also indicate ways in which knowledge can or must be acquired. Differences in the underlying truth theory lead to problems in appreciating why a piece of research may be considered valuable. This leads us to the second problem of knowledge, the question when a belief is justified.

3.2 Empiricism

Among the different ways of acquiring knowledge and defending the claim for truth, the most prominent one is probably empiricism. Empiricism can be defined as the "doctrine that experience rather than reason is the source of our knowledge of the world" (Morick, 1980 p. 1; cf. Gergen, 1999). Empiricism is the traditional epistemology of the natural sciences (Ciborra, 2002) where its adherents usually search for causal relationships. Empiricists trys to discover the laws governing reality and use a hypothetico-deductive approach (Vitalari, 1985). Empiricists set up hypotheses which they then try to prove of falsify (Popper, 1980). The ultimate aim of empiricist research is to be able to make well-founded predictions (Orlikowski & Baroudi, 1991; Westland, 2004)

Empiricism is closely associated with several assumption about the nature of scientific inquiry. First, it holds that observation is objective (Klein & Myers, 1999) and value-free (Walsham, 1995). It is also seen as a universally valid approach to knowledge which means that it is often associated with calls for a unity of science which would include the natural sciences as well as arts, humanities, and social sciences. Objectivity can be assured through an observer who is detached from the object of observation and who does not interfere (Introna, 1997; Yin, 2003). An important ingredient to this of approach to academic inquiry is a certain kind of detached and aloof rationality which is interested in relationships without being intimately involved in them (Wilson, 2003).

10 Chapter #

As a reaction to the perceived weaknesses of empiricism, which include the problem of the possibility of objectivity in social science, the question of appropriateness of empirical observation of humans, the alleged circularity of empiricism, the complexities of the notion of causality, a resistance to the underlying rationality, and other problems, other epistemological approaches have been developed. The most frequently quoted alternative to empiricism in philosophy is rationalism, which is the doctrine the reason, instead of sensation, is the foundation of knowledge.

3.3 Phenomenology

Phenomenology is an ambiguous term because it can refer to a general first-person description of human experience or, more specifically, to a philosophical method for analysing consciousness developed by Edmund Husserl (Beavers, 2002). The term has been used by Kant and Hegel, but Husserl redefined it in reaction to the detached academic discussion in philosophy in the 19th century. Heidegger, the possibly most important phenomenologist, defines the term "phenomenon" using its Greek etymology as "that which shows itself in itself, the manifest" (Heidegger, 1993 p. 28; cf. Moran, 2000)

Heidegger sees phenomenology as an ontology but it can also be understood as an epistemology. The central idea of phenomenology is that the world is opened up by consciousness. Every perception is a conscious act. Phenomena are given to consciousness and phenomenology tries to go back to the things themselves. These things are not objectively given things, but rather the content of consciousness (Lyotard, 1993). The phenomenologist tries to bracket out the non-essential aspects of perception to end up with the essence of the phenomenon.

In phenomenology the essences of the objects of research cannot be divided from the subject who researches them. The classical subject-object dichotomy of empiricism is not valid here (Moran 2000). An important aspect emphasised by Heidegger is that the subject of perception is never an independent entity but that it is a human with all the lived experiences and background, what he calls being-in-the-world, and for which he uses the term "Dasein". Dasein has to contend with the realities of human existence. It is embodied, it faces death, it is lonely and at the same time subjected to fashions, to the One (das Man) as Heidegger calls it (Dreyfus, 1993; Introna, 1997; Capurro & Pingel, 2002; Stuart, 2002; Introna & Whittaker, 2003). Since humans cannot live a detached and objective existence, they live in their own, partly idiosyncratic world, which Husserl calls life-world. The life-world is the strange thing that disintegrates before our eyes. It is the horizon within which we always move (Habermas, 1985; Stahl, 2003b).

#. Positivism or Non-Positivism - Tertium Non Datur 11

Phenomenology is thus a way of achieving knowledge and can be seen as an epistemology. It differs essentially from empiricism and is based on completely different assumption what knowledge is and how it can be acquired. But it is not the only possible alternative to empiricism.

3.4 Hermeneutics

Hermeneutics is another alternative to empiricism as a way of acquiring knowledge. Etymologically it is derived from the Greek word for "to interpret" (Hirschheim & Klein, 1989). The original purpose of hermeneutics was the understanding of religious texts, more specifically of the bible. It has developed into a general approach to the understanding of texts. The underlying problem is that every reader of a text has a different understanding of that text depending on his or her own experiences and life-world. This understanding differs from the understanding of the author. Originally, hermeneutics tried to find ways of determining the true sense of the text as intended by the author (or God). Hermeneutics has moved away from the idea of such a "correct" understanding and has expanded into the art of understanding all communication, not just written text.

One important aspect of contemporary hermeneutics is the hermeneutic circle. The idea behind this is that there is a circular relationship between the prior knowledge of a recipient of a text and her understanding of the same text. A text can only be read if the reader has a general understanding of its content but this understanding will be modified through the reading of the text (cf. Gadamer, 1990).

The current version of hermeneutics was explicitly developed to counter the natural science approach to humanities and social sciences. The opposition to natural sciences can best be demonstrated by looking at a pair of concepts associated with the German words erklären und verstehen (Hausman, 1994). Erklären, literally "to explain" refers to the natural sciences where causal relations can be established which can be used to explain phenomena. Such causal explanation are not useful in the humanities and social sciences because they negate the ability of agents to act. An explanation of human actions is thus not an application of natural laws but rather a description of humans that allows the reader to understand what the agent did and why she did it. This is what verstehen, literally "to understand" will achieve. Hermeneutics aims at facilitating this understanding. In hermeneutics there can be no unity of sciences. Social and natural sciences have different research objects and thus need different epistemologies (cf. Ricoeur, 1982; Mill, 1994).

According to this description of hermeneutics, the role of the researcher must be different from that in empiricism. The researcher cannot be detached

12 Chapter # and needs to admit that his or her understanding of the situation affects the outcome of the research (Myers & Avison, 2002).

A final remark on the relationship between hermeneutics and phenomenology: In its current form hermeneutics has been shaped by phenomenology. The most important hermeneutic philosophers, among them Gadamer and Ricoeur, were strongly influenced by phenomenology. If the phenomenon in question is a social one, as is typically the case in IS research, then a phenomenological researcher needs to acquire an understanding of the social exchange that constitutes it. For this, the researcher must apply hermeneutic means. He or she must follow the hermeneutic circle by starting with a given understanding, engaging with the phenomenon and thereby changing the initial understanding (cf. Boland, 1985).

3.5 Positivist and Non-Positivist Epistemology

Empiricism, phenomenology, and hermeneutics are the most important epistemologies in current IS research. In the context of this paper it is interesting to ask what their relationship to positivism and its alternatives is. There are tendencies that link ontologies and epistemologies. Positivism typically goes with empiricism whereas non-positivist approaches such as interpretivism tend to use hermeneutics or phenomenology as means to acquire knowledge. The question is whether these are necessary or contingent relationships. This paper will argue that these relationships are not necessary but that a change in the typical configurations requires rethinking the meaning of concepts.

There is, for example, no reason why a positivist should not use hermeneutics. The history of hermeneutics is actually based on a positivist understanding where one real and existing God put His thoughts to paper (using human instruments) and these words had one right meaning. Hermeneutics can be used as a tool to find out what this correct meaning of a divine text is. This meaning of "hermeneutics" is different from the post-Heideggerian hermeneutics described above. Conversely, an interpretivist can use an empiricist approach to research and try to observe reality and find objective truth. In this case, however, "objective truth" will mean something different from the positivist use of the term. It cannot be a correct description of an independent reality but must be something different, for example a validity claim that is not disputed (cf. Habermas, 1981).

While these questions are discussed frequently in IS research, particularly with regards to methodology (cf. next section), an equally important but generally not discussed problem is that of the relationship of ontology with truth theory. This is important because the assumed truth theory will

#. Positivism or Non-Positivism - Tertium Non Datur 13 determine what can count as valid research results. The dichotomy of positivism and non-positivism is mirrored more clearly with respect to truth theories. A correspondence theory of truth is only valid if there is a detached reality which one can describe correctly. Similarly, a consensus theory of truth is highly adequate to a constructionist ontology but holds no merits in an objective world. The coherence theory will play a role in the mathematical modelling, which is again best suited to positivism. Pragmatic truth theories, finally, can be appropriate for different ontologies but their meaning, the question what works, can reflect the ontology in different ways. For the positivist, a proposition is successful because it describes the world adequately, for the construcionist because it fulfils the criterion of being successfully constructed.

4. METHODOLOGY

While epistemology deals with the question what knowledge is, methodology asks how valid knowledge can be acquired. Methodology is thus the study of methods (Mingers, 2001), and it analyses the different methods used in research. There are numerous attempts to collect and classify research methods. Jenkins (1985), for example, identifies thirteen. The most important divide between methods is that between quantitative and qualitative methods. There has been an intensive discussion between proponents of the two sides in IS research for at least the last 20 years. I will not recount this discussion here.

The interesting question in this paper is whether research methodologies have a clearly defined relationship to the ontology upon which the research is based. Again, there are typical combinations between methods, epistemologies and ontologies, which are not always necessary. Positivists using a correspondence theory of truth and an empiricist epistemology will often use quantitative methods. However, there is no fundamental reason why they could not use qualitative methods (Urquhart, 2001). In order to find out the reality of a social phenomenon and to describe it as it objectively is, it may be helpful to observe agents or interview them, to write "realist tales" (van der Blonk, 2003). On the other hand, there is the typical combination of constructionist ontology, consensus theory of truth, hermeneutic / phenomenological epistemology and qualitative methods. Again, there is on a priori reason, however, why quantitative methods should not be used here. Numbers and statistics can be seen as ways of clarifying meanings and shared realities (Miranda & Saunders, 2003).

A related question of interest is whether different research methods can be combined, which is usually discussed under the heading of "multi-

14 Chapter # method" research or "pluralist research" (Landry & Banville, 1992; Sawyer, 2001; Brooke, 2002). This is a question on which the discussion of positivism versus non-positivism has a profound influence and to which we will return later.

Due to space constraints I will have to leave the questions of methodology at this rather superficial level. The reader should remember that there are typical methodological predispositions depending on the underlying ontology but no strictly necessary relationships.

5. ONTOLOGICAL SYNCRETISM IN IS RESEARCH

Syncretism is the "combination of different beliefs, the attempted combination of different systems of philosophical or religious belief or practice" (Encarta, 1999, p. 1893). It is usually regarded with scepticism by adherents of a given philosophy or religion. The fundamental question concerning syncretism is always whether the combined belief systems are compatible. This paper has argued that positivism and non-positivism are mutually exclusive and a syncretistic approach therefore cannot be acceptable. The consequent questions for this paper are: 1. is there indeed a philosophical syncretism to be observed in IS research? and 2. if so, how can it be explained?

5.1 The Existence of Philosophical Syncretism in IS Research

Analysing examples of philosophical syncretism in IS research is made difficult by the fact that most scholars follow their own definition of philosophical terms. An added difficulty is that some concepts seem to imply a mixture of positivism and non-positivism, such as multi-method research (Hirschheim, 1985; Cavaye, 1996), critical realism (Mingers, 2001; Mingers 2001b), but that the fundamental ontological issues are rarely spelt out.

The typical form of ontological syncretism in IS research is the assurance that positivism and non-positivism (usually interpretivism) can peacefully coexist and pose no threat to one another. If this paper is correct in arguing that positivism and non-positivism are logically mutually exclusive, then this constitutes syncretism. One typical example of this is the interpretivist researcher who wants to promote understanding for her research approach but who is careful not to offend positivist researchers by insisting that both approaches are valuable. "We must clearly state that it is not our intention to replace the positivist perspective with critical or interpretive ones"

#. Positivism or Non-Positivism - Tertium Non Datur 15 (Orlikowski & Baroudi, 1991 p. 24). Others see a more complex relationship where positivist and non-positivist research enters into some kind of dialectical process whereby higher level knowledge is produced (Klein, Hirschheim & Nissen, 1991). There are few examples of actually mixing positivist and non-positivist research approaches and contrasting the results of doing research on the basis of different ontologies and their resulting choices of epistemology and methodology (cf. Trauth & Jessup 2000). Nevertheless, most non-positivists seem to imply that positivism and non-positivism can coexist (Lee, 1991; 1994). The frequency of these syncretistic approaches allows Walsham (1995b) to identify four rhetorical figures used in the literature to justify the syncretistic approach.

5.2 Reasons for Syncretism

If syncretism between positivism and non-positivism is wide-spread in IS research despite the fact that the two approaches are based on contradictory ontologies, then the question must be why researchers try to combine the two. There are several reasons for this. The most important ones are lack of clarity of the concepts, confusion of the levels, research interests, metaphysical convictions, and the history and politics of the IS discipline.

The possibly most wide-spread reason for mixing positivism and non-positivism is the lack of clarity of the terms. This paper has made the point that we are looking at ontological positions which are associated with epistemological, methodological and arguably other research-relevant aspects. One should concede, however, that there is no unanimity on the exact limits and definition of the concept of positivism, and less so on its alternatives. The situation is worsened by the used of the term "paradigm". For Kuhn (1996) a paradigm is a set of assumptions that scientists agree on and that sets the frame for scientific research. This theory is interesting and fruitful because it can explain fundamental changes in the development of science. It is widely used in social sciences and information systems research (Burrell & Morgan, 1979; Chua, 1986, Hirschhein & Klein, 1989; Orlikowski & Baroudi, 1991; Goles & Hirschheim, 2000; Petter & Gallivan, 2004) where it has taken on a slightly different meaning. When IS researchers speak of paradigms they mean some sort of worldview, a general understanding of the world. Paradigms encompass all the aspects discussed in this paper and often others, such as ethical stances, social order, or the relationship between theory and practice. The problem with this use of the word paradigm is that it makes it very difficult to delimit and define meanings and contrast different paradigms. Typically, we speak of the positivist, the interpretivist, and the critical paradigm. Since methodologies used in the positivist paradigm can be used in conjunction with

16 Chapter # methodologies associated in the interpretive paradigm, some authors conclude that the two can coexist, thereby (wrongly) implying that the underlying ontologies can also coexist.

The unclear use of the term paradigm is partly responsible for the next reason for syncretism, namely the confusion of levels. It was argued here that the ontological assumptions of positivism and non-positivism are mutually exclusive. However, the same is not necessarily true for the associated epistemologies and methodologies. Typical examples of this are given by Lee (1991; 1994) when he tries to integrate the positivist and interpretivist approaches to organisational research. This mistake is closely related to the equating of ontology and methodology, where positivism stands for a certain methodology (usually quantitative methods) and interpretivism for non-quantitative methods (Benbasat & Weber, 1996; Cavaye, 1996; Landy & Banville, 1992; Eisenhardt 1989; Weber 2004). In these cases, the dichotomy between positivism and non-positivism becomes indistinguishable from the dichotomy between quantitative and qualitative research. Since the latter pair does not actually constitute a contradiction, the (wrong) conclusion is that positivism and non-positivism are also compatible.

Another reason of syncretism is research interest. This, too, is based on the confusion of the different philosophical levels. The argument here states that one gets a richer picture of a research object by looking at it from several sides. Similarly, the application of positivist and non-positivist research approaches is supposed to help IS researchers understand their object of interest better. The two positions are seen as complementary rather than contradictory. One can help explain (erklären) whereas the other helps understand (verstehen) (Lee, 1994). They allow triangulation through different methods and thus improve understanding. On the basis of these convictions one can call for a pluralism of methods. This understandable argument leads to the untenable conclusion that one should also mix the different ontologies.

Such attitudes may partly be explained by the history and politics of the IS discipline, which determines the constraints and requirements that IS researchers are subject to. A look at the history of IS (or MIS) shows that the discipline has been established for 30 to 50 years (Hirschheim & Klein, 2003; Ward & Peppard, 1996). The established "reference disciplines" (Keen, 1991), such as computer sciences, management sciences, organisation sciences, or economics (Benbasat & Weber, 1996), tend to use positivist assumptions. Given the traditional strength of positivism in IS, researchers are under strong pressure to recognise it as valid in order to get their PhD recognised or their research published (Baskerville, 2001). Then there are the politics of the IS discipline which has to survive among

#. Positivism or Non-Positivism - Tertium Non Datur 17 competing academic disciplines and which, according to some, lacks a recognisable core and definition (Benbasat & Zmud, 2003). It has been called a "fragmented adhocracy" (Landry & Banville, 1992) which does not bode well if it wants to survive among the other positivist subjects. Independent of the truth and usefulness of such statements, they serve to exert pressure on academics not to be seen as divisive. As a result, history and politics of the discipline combine to strongly pressure non-positivist researchers to accept positivist approaches even if they agree with the argument of this paper.

All of the above arguments allowed for syncretism by shirking the issue of the ontological contradiction between positivism and its alternatives. There are, however, some arguments that directly oppose my thesis of the contradictory nature of positivism and its alternatives. The most problematic of these is the denial that there is a difference between positivism and non-positivism. This seems to be grounded in the positivists’ fundamental inability to understand that there are alternatives to their realist ontology. Weber (2004, v) exemplifies this position when he incredulously exclaims that "[… ] surely some kind of reality exists beyond our perceptions of it!" and posits that it is inconceivable that anybody might contradict this statement. As proof he offers the certain reality of death faced by everybody who jumps out of his office window. He neglects to see that, for the phenomenologist, death is always-mine (jemeinig, as Heidegger (1992) would have said) and thus the epitome of an idiosyncratic and thus non-objective experience. On a possibly more accessible level, the problem has also been discussed by Grint & Woolgar (1997) under the heading of "what is social about being shot?"

There are also more sophisticated attempts to reconcile the ontological differences. These can be based on post-positivism, postmodernism, critical realism, Heglian dialectics, pragmatism or others (cf. Petter & Gallivan, 2004; Hirschheim, 1985; Varey, Wood-Harper & Wood, 2002; Achterberg, van Es & Heng, 1991; Goles & Hirschheim, 2000). There is no space here to discuss these but, briefly, there seems to be no way they can avoid the dichotomy between an observer-independent and an observer-dependent reality. At best they open up a new meta-theoretical approach to reality, which typically depends on the observer and is thus non-positivist.

6. CONSEQUENCES OF THE DICHOTOMY OF POSITIVISM AND NON-POSITIVSM

If the ontological foundations of positivism and non-positivism are not compatible; if, at the same time, IS researchers tend to mix them, then one

18 Chapter # should ask what consequences it would have to change this practice and own up to the unbridgeable dichotomy of ontologies. Is Weber (2004, p. vi) possibly right when he says that it "makes no difference to the fundamental goals [… ]" of researchers? This section will briefly look at implications in the area of research, politics, and ethics.

6.1 Research Implications

There is no algorithmic way of doing research according to one's ontological position. This paper does not support the thesis of the incommensurability of different methods (quantitative vs. qualitative). The main conclusion to be drawn is that the combination of different methods, epistemologies, and ontologies has to be justified in every single instance. It may be completely acceptable and logically stringent to use quantitative methods from an interpretivist viewpoint or to use semi-structured interviews from a positivist viewpoint. At the same time, the same research method will mean different things depending on one's ontology. A positivist doing interviews will expect to find social reality as it is, whereas the constructionist will be part of the collective construction of the relevant reality. Two researchers using the exact same approach and getting the same results may thus come to opposing conclusions, based on their ontological underpinnings.

A related but more difficult question is that of the compatibility of epistemologies. Empiricist and hermeneutic / phenomenological approaches seem to be more difficult to combine than quantitative and qualitative methods. This is probably based on their greater closeness to the underlying ontology. One problem results from the fact that most interpretive IS research is based on the collection of empirical evidence. That means that empirical research is not necessarily empiricist. It also means that non-empiricist researchers doing empirical research should spell out why they believe that this will help them. In the light of a non-positivist ontology and a non-empiricist epistemology it is not immediately obvious that empirical research is superior to other kinds, such as philosophical or conceptual research (Stahl, 2003)

The main research implications of the positivism - non-positivism divide is that there are individual and collective responsibilities with regards to the combinations of methodologies, epistemologies and ontologies (Robey, 1996). A simple pick-and-choose approach guided by considerations of expediency cannot be acceptable. That means that the individual researcher should be clear about these questions and should address them in his or her research design. On the collective side, the discipline as represented by

#. Positivism or Non-Positivism - Tertium Non Datur 19 conferences or journals, chairs, reviewers, or editors needs to make sure that these questions are properly and satisfactorily reflected.

A related aspect, leading us to the politics of IS research, are standards of research quality, validity, rigour, or relevance. These much-debated problems will not be compatible for positivists and non-positivists. While they are in some instances related to specific methods, the underlying ontology determines whether a certain piece of research is acceptable. A central point here will be the truth theory used to assess these issues. These are very closely linked to the underlying ontology and they are currently rarely reflected in detail.

For readers more familiar with the discussion of ontology in IS rather than general philosophical ontology, a question of interest may be what the above debate has to do with computational or IS ontology. I believe that any discussion of specific ontologies requires a foundation in philosophical ontology. There are numerous issues of philosophical ontology that this paper has not addressed. But the question whether reality is independent of the observer is pertinent to all ontologies. If Kishore, Sharman & Ramesh (2004) are right and there are different "universes of discourse" which establish different ontologies, then the question is what ontological status these ontologies have. The existence of multiple ontologies seems to imply that there is not one reality as assumed by positivism but a number of them. These are then presumably dependent on the observer and the situation she finds herself in. Computational ontologies, which have as a main purpose to provide a taxonomy that will allow dealing with computational entities are thus non-positivist (even if they are typically not explicitly or intentionally linked to a philosophical ontology). If this is true, we need to ask where the influence of the observer can be seen in the resulting taxonometic systems. These should play a role in the verification, validation, and assessment of ontologies. A cursory glance at the literature suggests that this is not the case. Computational ontologies seem to claim objective validity and thus aim for a positivist description of reality (Kishore, Sharman & Ramesh, 2004b). Again, if this is true, then there may be a potential contradiction in the ontological foundations of regional ontologies which would need to be addressed.

6.2 Political Implications

Research does not take place in a vacuum (cf. Lee, 2001) but is embedded in social systems where politics play a great role. This should be quite obvious particularly for non-positivist researchers who recognise the social construction of reality. Yet, non-positivist research politics are much less visible or successful than their positivist counterpart. Representatives of

20 Chapter # the positivist view are openly trying to set the agenda of IS research according to their ideas. They try to define legitimate research subjects like the IT artefact and to impose an "identity" on the field (Benbasat & Zmud, 2003). They promote certain research methodologies and theories, favouring formal and mathematical methods, which are typically more useful to their ontology than to others (Weber, 2003).

While these attempts to promote the positivist agenda have created a lively debate (see the "Core of IS" debate in the Communications of the AIS), no concerted non-positivist action is visible. The reasons for this are manifold, but among the most important ones one can find the historical prevalence of positivism (Walsham, 1995b; Trauth, 2001) and the lack of coherence among non-positivist. This paper may help alleviate the problem by offering the idea that the overarching characteristic of non-positivist research is its ontological belief of the dependence of reality on the observer.

6.3 Ethical Implications

A final important implication of recognising the incommensurability of positivism and non-positivism has to do with the ethics of research, and more specifically with how humans are perceived in research. Positivism requires ethical behaviour by the researcher (Hausman & McPherson, 1994) and it can even be described as an ethically motivated endeavour that promised to develop society (Wynn, 2001) and "limitless progress" (Chomsky, 1998 p. 128). It would thus be wrong to see positivism as fundamentally "un-ethical" but in effect it develops worrying ethical consequences. These are the result of the perceived possibility of distinguishing clearly between research object and subject and between objective description and subjective evaluation. This allows positivists to argue that they are detached observers which negates the necessity to become involved and therefore roots for the status quo (Orlikowski & Baroudi, 1991)

A positivist ontology suggests that all things exist in some sort of objective universe and this includes human beings. This, combined with methodological individualism, can create a disposition to treat humans as objects. Treating humans as objects means that one can treat them as means rather than ends, thus violating Kant’s (1995, p. BA 67) famous version of the Categorical Imperative according to which humans should never be treated as means. Or to put it in more contemporary words: "Such research may end up by recommending most people to be handled like billiard balls" (Nissen, 1985 p. 40).

#. Positivism or Non-Positivism - Tertium Non Datur 21 7. CONCLUSION: A TIME FOR TOLERANCE?

The paper has argued that positivism and non-positivism as ontological positions are irreconcilable. There are epistemological and methodological consequences of the ontological divide, but these are less clear-cut than the ontological opposition. The paper has outlined some of the implications a recognition of the divide as they would arise for research, politics, and ethics.

Given the philosophical and conceptual nature of the argument, it is impossible to "prove" it wrong by using contradicting empirical data. Ontology is not subject to empirical investigation because any empirical research must be based on an ontology which it cannot prove wrong because it determines which phenomena can be observed. The main area of contention will lie in the use of the concepts themselves. One potential weakness of the paper is that the general use of the concepts such as positivism, interpretivism, empiricism, etc. does not always follow the definitions offered here. The paper has tried to show that these definitions are tenable and well-grounded in the literature but it is also true that some authors use them differently. The answer to such a claim would be that a different use of terms does not affect the content of the argument. A researcher must base his or her research on an understanding of the nature of reality. He or she will be faced with the dichotomous choice of an observer-independent or observer-dependent reality. The two sides cannot be bridged for logical reasons. Resulting from or at least affected by the choice, the researcher will use certain epistemologies or methodologies. The argument thus stands as it is, even if one does not agree with the definitions of "positivism" or "non-positivism".

But even if this problem exists as suggested in the paper, do IS researchers or practitioners need to worry about it? Are there conclusions that need to be drawn or consequences that arise from it? In the light of the current debate about the core and definition of the field of information systems, the answer has to be affirmative. For many everyday purposes, researchers may be able to rely on their ontology in use and not worry about the underlying problems. This is at least true for those who use the predominant ontology of positivism. Researchers who prefer a different approach are put more often in a position where they have to justify their ontology. But in the overall climate of scarcity of resources (university chairs, research positions, studentships, funding, etc.) we need to consider which criteria constitute valuable work in IS and these criteria are inherently dependent on the underlying ontology. If the paper is right and positivism and non-positivism are not commensurable, then the discipline of IS should ask itself how it should view and possibly compare research from these two

22 Chapter # traditions. A radical solution would be to say that they are so fundamentally different that they do not in fact belong to the same discipline and that we therefore should establish two different disciplines. For the political reasons mentioned above, this may not be a desirable position. A less radical solution would be to continue to coexist under the roof of "information systems" but to generally ignore each other. One can probably argue that this is the current state of the discipline. But if we want to coexist, will we be able to tolerate one another well enough to remain under the same roof despite competition for the same resources?

Finally, there is the crucial question of the overall context of this discussion. This paper has concentrated on philosophical arguments in the context of IS research. The question of ontology is by no means confined to research or to a specific discipline. On the contrary, it is a central assumption in our individual and collective world-views. It strongly influences questions of politics, social distribution, of war and peace. The problem of tolerance between positivist and non-positivist IS researchers would thus have to be viewed in the context of tolerance between positivists and non-positivists in general. To stay within the religious terminology of the title of this paper, tolerance might be translated into an ecumenical approach. A brief look at the history of religious wars and the ensuing ecumenical movements suggests that peaceful coexistence of contradictory beliefs may be possible but tends to be very fragile. It is also only possible if the different sides desire it deeply and are willing to accept the other position as possibly misguided but legitimate. It is open to debate whether these conditions are met in contemporary IS research.

8. REFERENCES

Achterberg, J. S., van Es, G. A. & Heng, M. S. H., 1991, Information systems research in the postmodern period. In: Information Systems Research: Contemporary Approaches & Emergent Traditions, Nissen, H. -E.; Klein, H. K. & Hirschheim, R., eds., North Holland, Amsterdam, pp. 281 - 294

Adam, A., 2001, Gender and Computer Ethics. In: Readings in Cyberethics, Spinello, R. A. & Tavani, H. T., eds., Jones and Bartlett, Sudbury, Massachusetts, pp. 63 - 76

Baskerville, R., 2001, Conducting Action Research: High Risk and High Reward in Theory and Practice. In: Qualitative Research in IS: Issues and Trends, Trauth, E., ed., Idea Group Publishing, Hershey, pp. 192 - 217

Beavers, A. F., 2002, Phenomenology and Artificial Intelligence, Metaphilosophy (33:1/2), Special Issue: Cyberphilosophy: The Intersection of Philosophy and Computing. Edited by J.H. Moor and T.W. Bynum, 70 - 82

Benbasat, I. & Weber, R., 1996, Research Commentary: Rethinking “Diversity” in Information Systems Research. Information Systems Research (7:4), 389 - 399

Benbasat, I. & Zmud, R. W., 2003, The Identity Crisis Within the IS Discipline: Defending and Communicating the Discipline's Core Properties, MIS Quarterly, (27:2), 183 - 194

#. Positivism or Non-Positivism - Tertium Non Datur 23 Benbasat, I. & Zmud, R. W., 1999, Empirical Research in Information Systems: The Practice

of Relevance. MIS Quarterly, (23:1), 3 - 16 Boland, R. J. jr., 1985, Phenomenology: A Preferred Approach to Research on Information

Systems. In: Research Methods in Information Systems (IFIP 8.2 Proceedings), Mumford, E.; Hirschheim, R.; Fitzgerald, G. & Wood-Harper, T. (eds), 1985,, North-Holland, Amsterdam, pp. 193 - 201

Brooke, C., 2002, Critical Perspectives on Information Systems: An Impression of the Research Landscape. Journal of Information Technology 17, 271 - 283

Burrell, G. & Morgan, G., 1979, Sociological Paradigms and Organizational Analysis, Heinemann, London

Capurro, R. & Pingel, C., 2002, Ethical Issues of Online Communication Research, Ethics and Information Technology (4:3,, Special Issue on Internet Research Ethics, edited by C. Ess, 189 - 194

Cavaye, A. L. M., 1996, Case Study Research: A Multi-Faceted Research Approach for IS, Information Systems Journal 6, 227 - 242

Chomsky, N., 1998, Reflections on Language (orig. 1975,. Reprinted in: On Language), Chomsky, N., The New Press, New York

Chua, W. F., 1986, Radical Developments in Accounting Thought. The Accounting Review (61:4), 601 - 632

Ciborra, C., 2002, The Labyrinths of Information - Challenging the Wisdom of Systems, Oxford University Press, Oxford

Darke, P.; Shanks, G. & Broadbent, M., 1998, Successfully Completing Case Study Research: Combining Rigour, Relevance and Pragmatism. Information Systems Journal (8), 273 - 289

Davies, L. J., 1991, Researching the Organizational Culture Context of Information Systems Strategy: A Case Study of the British Army. In: Information Systems Research: Contemporary Approaches & Emergent Traditions, Nissen, H. -E.; Klein, H. K. & Hirschheim, R., eds., North Holland, Amsterdam, pp. 145 - 167

Dreyfus, H. L., 1993, What Computers Still Can’t Do, MIT Press, Cambridge, Massachusetts / London

Encarta, 1999, Encarta World English Dictionary, Bloomsbury, London Eisenhardt, K. M., 1989, Building Theories from Case Study Research, Academy of

Management Review, (14:4), 632 - 550 Feyerabend, P. K., 1980, How to Be a Good Empiricist - A Plea for Tolerance in Matters

Epistemological. In: Challenges to Empiricism, Morick, Harold, ed., Methuen, London, pp. 164 - 193

Fitzgerald, B. & Howcroft, D., 1998, Towards Dissolution of the IS Research Debate: from Polarization to Polarity, Journal of Information Technology 13 (Special Issue on Interpretive Research in Information Systems, edited by M. Myers and G. Walsham), 313 - 326

Friedman, M., 1994, The Methodology of Positive Economics. In: The Philosophy of Economics: An Anthology, Hausman, D. M., ed., pp. 180 - 213, 2nd edition Cambridge University Press , Cambridge

Gadamer, H. -G., 1990, Wahrheit und Methode - Grundzüge einer philosophischen Hermeneutik. 6th edition, J.C.B. Mohr (Paul Siebeck), Tübingen

Garson, J., 2003, Modal Logic. In: The Stanford Encyclopedia of Philosophy (Winter 2003 Edition), Zalta, Edward N., ed., URL = <http://plato.stanford.edu/archives/win2003/entries/logic-modal/>, [accessed 09.07.2004]

Gergen, K. J., 1999, An Invitation to Social Construction, Sage, London

24 Chapter # Glasersfeld, von, E., 2000, Konstruktion der Wirklichkeit und des Begriffs der Objektivität.

In: Einführung in den Konstruktivismus, Carl Friedrich von Siemens Stiftung, ed., 5th edition, Piper, München, pp. 9 - 40

Goles, T. & Hirschheim, R., 2000, The Paradigm is Dead, the Paradigm is Dead... Long Live the Paradigm: the Legacy of Burrell and Morgan, Omega, (28:3), 249 - 268

Grint, K. & Woolgar, S., 1997, The Machine at Work: Technology, Work, and Organization, Blackwell, Cambridge

Habermas, J., 1991, Erläuterungen zur Diskursethik, Suhrkamp, Frankfurt a. M. Habermas, J., 1985, Die neue Unübersichtlichkeit, Suhrkamp, Frankfurt a. M. Habermas, J., 1981, Theorie des kommunikativen Handelns, Suhrkamp Verlag, Frankfurt a.

M. Habermas, J., 1974, Erkenntnis und Interesse, Suhrkamp, Frankfurt a. M. Hajek, P., 2002, Fuzzy Logic. In: The Stanford Encyclopedia of Philosophy (Fall 2002

Edition), Zalta, E.N., ed., URL = <http://plato.stanford.edu/archives/fall2002/entries/logic-fuzzy/> [accessed 09.07.2004]

Hausman, D. M., ed., 1994, The Philosophy of Economics: An Anthology, 2nd edition, Cambridge University Press, Cambridge

Hausman, D. M & McPherson, M. S., 1994, Economics, rationality, and ethics. In: The Philosophy of Economics: An Anthology, 2nd edition, Hausman, D. M., ed., Cambridge University Press, Cambridge, pp. 252 - 277

Heidegger, M., 1993, Sein und Zeit, 17th edition, Max Niemeyer Verlag, Tübingen Hirschheim, R. A., 1985, Information Systems Epistemology: An Historical Perspective. In:

Research Methods in Information Systems (IFIP 8.2 Proceedings), Mumford, E.; Hirschheim, R.; Fitzgerald, G. & Wood-Harper, T. (eds), North-Holland, Amsterdam, pp. 13 - 36

Hirschheim, R. & Klein, H. K., 2003, Crisis in the IS Field? A Critical Reflection on the State of the Discipline, Journal of the Association for Information Systems (4:5), 237 - 293

Hirschheim, R. & Klein, H. K., 1989, Four Paradigms of Information Systems Development. Communications of the ACM (32:10), 1199 - 1216

Hollis, M., 1994, The Philosophy of Social Science: an Introduction. Cambridge University Press, Cambridge

Iivari, J.; Hirschheim, R. & Klein, H. K., 1998, A Paradigmatic Analysis Contrasting Information Systems Development Approaches and Methodologies, Information Systems Research (9:2), 164 - 193

Introna, L., 1997, Management, Information and Power: A narrative of the involved manager, MacMillan, London

Introna, L. D. & Whittaker, L., 2003, The Phenomenology of Information Systems Evaluation: Overcoming the Subject/Object Dualism. In: Global and Organizational Discourse About Information Technology, Wynn, E.; Whitley, E.; Myers, M. D. & DeGross, J., eds., Kluwer Academic Publishers, Dordrecht, pp. 156 - 175

Jenkins, Milton A., 1985, Research Methodologies and MIS Research. In: Research Methods in Information Systems (IFIP 8.2 Proceedings), Mumford, E.; Hirschheim, R.; Fitzgerald, G. & Wood-Harper, T. (eds), North-Holland, Amsterdam, pp. 103 - 117

Jönsson, S., 1991, Action Research. In: Information Systems Research: Contemporary Approaches & Emergent Traditions, Nissen, H. -E.; Klein, H. K. & Hirschheim, R., eds., North Holland, Amsterdam, pp. 371 - 396

Kant, I., 1995, Kritik der praktischen Vernunft, Grundlegung zur Metaphysik der Sitten,: Suhrkamp, Frankfurt a. M

Keen, P. G. W., 1991, Relevance and Rigor in Information Systems Research: Improving Quality, Confidence, Cohesion and Impact. In: Systems Research: Contemporary

#. Positivism or Non-Positivism - Tertium Non Datur 25

Approaches & Emergent Traditions, Nissen, Hans-Erik; Klein, Heinz K. & Hirschheim, Rudy, eds., North Holland, Amsterdam, pp. 27 - 49

Khlentzos, Drew, 2004,: Naturalistic Realism and the Antirealist Challenge. Cambridge, Massachusetts: The MIT Press

Kishore, Rajiv; Sharman, Raj & Ramesh, Ram, 2004, Computational Ontologies and Information Systems: I. Foundations. In: Communications of the Association for Information Systems 14, 158 - 183

Kishore, Rajiv; Sharman, Raj & Ramesh, Ram, 2004b, Computational Ontologies and Information Systems: II. Formal Specifications. In: Communications of the Association for Information Systems 14, 184 - 205

Klein, H. K.; Hirschheim, R. & Nissen, H. -E., 1991, A Pluralist Perspective of the Information Systems Research Arena. In: Information Systems Research: Contemporary Approaches & Emergent Traditions, Nissen, H. -E.; Klein, H. K. & Hirschheim, R., eds., pp. 1 - 17 North Holland, Amsterdam

Klein, H. K. & Myers, M. D., 2001, A Classification Scheme for Interpretive Research in Information Systems. In: Qualitative Research in IS: Issues and Trends, Trauth, Eileen, ed., pp. 218 - 239, Idea Group Publishing, Hershey

Klein, H. K. & Myers, M. D., 1999, A Set of Principles for Conducting and Evaluating Interpretive Field Studies in Information Systems. MIS Quarterly, (23:1), 67 - 94

Kuhn, T. S., 1996, The Structure of Scientific Revolutions, 3rd edition, The University of Chicago Press, Chicago and London

Landry, M. & Banville, C., 1992, A Disciplined Methodological Pluralism for MIS Research. Accounting, Management & Information Technology, 2:2,, 77 - 92

Lee, A., 2004,: Thinking about Social Theory and Philosophy for Information Systems. In: Social Theory and Philosophy for Information Systems, Mingers, John & Willcocks, Leslie, eds., Chichester: Wiley, pp. 1 - 26

Lee, A., 2001, Challenges to Qualitative Researchers in IS. In: Qualitative Research in IS: Issues and Trends, Trauth, E., ed. Idea Group Publishing, Hershey, pp. 240 - 270

Lee, A., 1994, Electronic Mail as a Medium for Rich Communication: An Empirical Investigation Using Hermeneutic Interpretation. MIS Quarterly, (18:2), 143 - 157

Lee, A., 1991, Integrating Positivist and Interpretive Approaches to Organizational Research. Organization Science, (2:4), 342 - 365

Lee, A. S & Baskerville, R. L., 2003, Generalizing Generalizability in Information Systems Research. Information Systems Research, (14:3), 221 - 243

Lyotard, J. -F., 1993, Die Phänomenologie, Junius Verlag, Hamburg Miranda, S. M. & Saunders, C. S., 2003, The Social Construction of Meaning: An Alternative

Perspective on Information Sharing. Information Systems Research, (14:1), 87 - 106 McCarthy, T., 1992, Philosophy and Social Practice: Avoiding the Ethnocentric Predicament.

In: Philosophical Interventions in the Unfinished Project of Enlightenment, Honneth, A.; McCarthy, T.; Offe, C. & Wellmer, A., MIT Press, Cambridge, Massachusetts & London, pp. 241 - 260

Mingers, J., 2001, Combining IS Research Methods: Towards a Pluralist Methodology, Information Systems Research, (12:3), 240 - 259

Mingers, J., 2001b) Embodying Information Systems: the Contribution of Phenomenology, Information and Organization, (11:2), 103 - 128

Mill, J. S., 1994, On the definition and method of political economy. In: The Philosophy of Economics: An Anthology, Hausman, D. M., ed., 2nd edition, Cambridge University Press, Cambridge, pp 52 - 68

26 Chapter # Monod, E., 2002, For a Kantian Foundation of IS Research: Proposals for an Epistemological

Pluralism. In: Proceedings of the Eighth Americas Conference on Information Systems 2002, 1751 - 1759

Moran, D., 2000, Introduction to Phenomenology, Routledge, London / New York Morick, H., 1980, Introduction: The Critique of Contemporary Empiricism. In: Challenges to

Empiricism, Morick, Harold, ed., Methuen, London, pp. 1 - 25 Myers, M., D & Avison, D., 2002, An Introduction to Qualitative Research in Information

Systems. In: Qualitative Research in Information Systems: a Reader, Myers, M., D & Avison, D., eds., Sage, London, pp. 3 - 12

Mumford, E.; Hirschheim, R.; Fitzgerald, G. & Wood-Harper, T. (eds), 1985, Research Methods in Information Systems (IFIP 8.2 Proceedings), North-Holland, Amsterdam

Nissen, H. -E., 1985, Acquiring Knowledge of Information Systems - Research in a Methodological Quagmire. In: Research Methods in Information Systems (IFIP 8.2 Proceedings), Mumford, E.; Hirschheim, R.; Fitzgerald, G. & Wood-Harper, T. (eds), North-Holland, Amsterdam, pp. 39 - 51

OED (Oxford English Dictionary), 2004,, http://athens.oed.com/ [accessed 01.06.2004] Olaison, J., 1991, Pluralism or Positivistic Trivialism: Important Trends in Contemporary

Philosophy of Science. In: Information Systems Research: Contemporary Approaches & Emergent Traditions, Nissen, H. -E.; Klein, H. K. & Hirschheim, R., eds., Amsterdam: North Holland, pp. 235 - 264

Orlikowski, W. J. & Baroudi J. J., 1991, Studying Information Technology in Organizations: Research Approaches and Assumptions. Information Systems Research, (2:1), 1 - 28

Petter, S. C. & Gallivan, M. J., 2004, Toward a Framework for Classifying and Guiding Mixed Method Research in Information Systems. In: Proceedings of the 37th Annual Hawaii International Conference on Systems Sciences, Hawaii, January 5-8, 2004

Pettigrew, A., 1985, Contextualist Research and the Study of Organisational Change Processes. In: Research Methods in Information Systems (IFIP 8.2 Proceedings), Mumford, E.; Hirschheim, R.; Fitzgerald, G. & Wood-Harper, T. (eds), North-Holland, Amsterdam, pp. 53 - 78

Popper, K. R., 1980, Science: Conjectures and Refutations. In: Challenges to Empiricism, Morick, Harold, ed., Methuen, London, pp. 128 - 160

Quine, W. V. O., 1980, Two Dogmas of Empiricism. In: Challenges to Empiricism, Morick, H., ed., Methuen, London, pp. 46 - 70

Ricoeur, P., 1995, Le Juste, Editions Esprit, Paris Ricoeur, P., 1983, Temps et récit - 1. L'intrigue et le récit historique, Editions de Seuil, Paris Robey, D., 1996, Research Commentary: Diversity in Information Systems Research: Threat,

Promise, and Responsibility, Information Systems Research (7:4), 400 - 408 Rorty, R., 1991, Objectivity, Relativism and Truth, Cambridge University Press, Cambridge Sawyer, S., 2001, Analysis by Long Walk: Some Approaches to the Synthesis of Multiple

Sources of Evidence. In: Qualitative Research in IS: Issues and Trends, Trauth, E., ed., Idea Group Publishing, Hershey, pp. 163 - 190

Stahl, B. C., 2003, How We Invent What We Measure: A Constructionist Critique of the Empiricist Bias in IS Research In: Proceedings of the Ninth Americas Conference on Information Systems, Tampa, 04 to 06 August 2003, 2878 - 2884

Stahl, B. C., 2003b, When does a Computer Speak the Truth? The Problem of IT and Validity Claims. In: Global and Organizational Discourse About Information Technology, Wynn, E.; Whitley, E.; Myers, M. D. & DeGross, J., eds., Kluwer Academic Publishers, Dordrecht, pp. 91 - 107

Steup, M., 2001, The Analysis of Knowledge, The Stanford Encyclopedia of Philosophy (Spring 2001 Edition), Edward N. Zalta, ed., URL =

#. Positivism or Non-Positivism - Tertium Non Datur 27

http://plato.stanford.edu/archives/spr2001/entries/knowledge-analysis/, [accessed 01.06.2004]

Stuart, S., 2002, A Radical Notion of Embeddedness: A Logically Necessary Precondition for Agency and Self-Awareness. Metaphilosophy (33:1/2), Special Issue: Cyberphilosophy: The Intersection of Philosophy and Computing. Edited by J.H. Moor and T.W. Bynum, 98 - 109

Trauth, E., 2001, The Choice of Qualitative Method in IS Research. In: Qualitative Research in IS: Issues and Trends, Trauth, Eileen, ed., Idea Group Publishing, Hershey, pp. 1 - 19

Trauth, E. M & Jessup, L. M., 2000, Understanding Computer-Mediated Discussions: Positivist and Interpretive Analyses of Group Support System Use. MIS Quarterly, (24:1), 43 - 79

Tugendhat, E., 1992, Reflections on Philosophical Method from an Analytic Point of View. In: Philosophical Interventions in the Unfinished Project of Enlightenment, Honneth, A.; McCarthy, T.; Offe, C. & Wellmer, A, eds., MIT Press, Cambridge, Massachusetts & London, pp. 113 - 124

Urquhart, C., 2001, An Encounter with Grounded Theory: Tackling the Practical and Philosophical Problems. In: Qualitative Research in IS: Issues and Trends, Trauth, E., ed., Idea Group Publishing, Hershey, pp. 104 - 140

van de Ven, A. H., 1989, Nothing Is Quite So Practical as a Good Theory. Academy of Management Review, (14:4), 486 - 489

van der Blonk, H., 2003, Writing Case Studies in Information Systems Research. Journal of Information Technology 18, 45 - 52

Varey, R. J.; Wood-Harper, T. & Wood, B., 2002, A Theoretical Review of Management and Information Systems Using a Critical Communications Theory. Journal of Information Technology 17, 229 - 239

Visala, S., 1991, Broadening the Empirical Framework of Information Systems Research. In: Information Systems Research: Contemporary Approaches & Emergent Traditions, Nissen, H. -E.; Klein, H. K. & Hirschheim, R., eds., North Holland, Amsterdam, pp. 347 - 364

Vitalari, N. P., 1985, The Need for Longitudinal Designs in the Study of Computing Environments. In: Research Methods in Information Systems (IFIP 8.2 Proceedings), Mumford, E.; Hirschheim, R.; Fitzgerald, G. & Wood-Harper, T. (eds), North-Holland, Amsterdam, pp. 243 - 265

Walsham, G., 1995, Interpretive Case Studies in IS Research: Nature and Method. European Journal of Information Systems 4, 74 - 81

Walsham, G., 1995b) The Emergence of Interpretivism in IS Research, Information Systems Research (6:4), 376 - 394

Ward, J. & Peppard, J., 1996, Reconciling the IT/business relationship: a troubled marriage in need of guidance, Journal of Strategic Information Systems 5, 37 - 65

Watzlawik, P., ed., 2001, Die erfundene Wirklichkeit, 13th edition, Piper, München, Zürich Weber, R., 2004, The Rhetoric of Positivism Versus Interpretivism: A Personal View

(Editor's Comment), MIS Quarterly, (28:1), iii - xii Weber, R., 2003, Theoretically Speaking (Editor's Comment), MIS Quarterly, (27:3), iii - xii Weick, K. E., 1989, Theory Construction as Disciplined Imagination. Academy of

Management Review, (14:4), 516 - 531 Weizenbaum, J., 1976, Computer Power and Human Reason, W. H. Freeman and Company,

San Francisco Wellmer, A., 1986, Ethik und Dialog: Elemente des moralischen Urteils bei Kant und in der

Diskursethik, Suhrkamp, Frankfurt a. M.

28 Chapter # Westland, J. C., 2004, The IS Core XII: Authority, Dogma and Positive Science in

Information Systems Research. Communications of the Association for Information Systems 13, 136 - 157

Wilson, M, 2003, Rhetoric of Enrollment and Acts of Resistance: Information Technology as Text. In: Global and Organizational Discourse About Information Technology, Wynn, Eleanor; Whitley, Edgar; Myers, Michael D. & DeGross, Janice, eds., Kluwer Academic Publishers, Dordrecht, pp.225 - 248

Wittgenstein, L., 2001, Philosophical Investigations / Philosopische Untersuchungen (translated by G.E.M. Anscombe), 3rd edition, Blackwell, Oxford

Wittgenstein, L., 1963, Tractatus logico-philosophicus - Logisch-philosophische Abhandlungen, Suhrkamp, Frankfurt a. M.

Wynn, E., 2001, Möbius Transitions in the Dilemma of Legitimacy. In: Qualitative Research in IS: Issues and Trends, Trauth, E., ed., Idea Group Publishing, Hershey, pp. 20 - 44

Wynn, E.; Whitley, E.; Myers, M. D. & DeGross, J., eds., 2003, Global and Organizational Discourse About Information Technology: Ifip Tc8/Wg8.2 Working Conference on Global and Organizational Discourse About Information Technology, December 12-14, 2002, Barcelona, Spain, Kluwer Academic Publishers, Dordrecht

Yin, R. K., 2003, Case Study Research: Design and Methods, 3rd edition, Sage, Thousand Oaks / London / New Delhi